INSIGHT INTO SPATIAL ANALYSIS OF GPS BROADCAST ORBIT ERROR TOWARDS ORBITAL IMPROVEMENT

Broadcast orbits are compared against final orbit to get the error of broadcast orbit. The errors are analysed by presenting the error over space, especially longitude. The satellite trajectory is divided into three sector namely northern, southern, and transitional sectors. Spatial analysis show that the error is correlated with the latitude and longitude. Some consistency pattern can be observed from the distribution of the error in the spatial analysis. Standard deviation (SD) is used to quantify the consistency, providing more quantitative insights into the spatial analysis. Four patterns can be observed in the error distribution, namely consistency in northern and southern sector, consistency of transitional sector, changes after transitional sector, and correlation between ΔX component and ΔY component. The spatial analysis shows potential to be used in broadcast orbit error estimation and prediction. A model that uses this predicted broadcast orbit error as a correction will be designed in the future to improve the broadcast orbit accuracy.

Application of adaptive identification methods for refining parameters of radiation pressure models

Представлены две адаптивные модификации сигма-точечного фильтра Калмана с рекуррентным оцениванием ковариационных матриц шумов системы и измерений, на основе которых выполняется процедура параметрической идентификации нелинейных непрерывно-дискретных систем. Применение процедуры адаптивной параметрической идентификации позволило вычислить с достаточной точностью оценки параметров нескольких моделей радиационного давления солнечного излучения. Полученные результаты повысили качество прогнозирования траектории движения навигационного спутника Purpose. The paper considers the problem of estimation of unknown parameters for various models of solar radiation based on adaptive modifications of the unscented Kalman filter. The estimations of the obtained parameters are used both in solar radiation models and in construction of trajectory of a navigation satellite. Methodology. To solve the problem of parametric identification of stochastic nonlinear continuous-discrete systems, several adaptive modifications of the unscented Kalman filter are considered. The algorithms assume recurrent estimation of covariance matrices of system noise and measurements. The maximum likelihood method is used for parametric identification of stochastic nonlinear continuous-discrete systems. Adaptive modifications of the unscented Kalman filter are used in the construction of the identification criterion. Estimates of unknown parameters of various solar radiation models are found for the movement for the navigation satellite model as an example. The satellite orbital movement forecast is made. Finding and value. The application of the adaptive parametric identification procedure allows calculating the estimates for the parameters of several models of the solar radiation with sufficient accuracy. The obtained results lead to significant improvement of quality of the prediction for satellite trajectory

Averaging Bias Correction for the Future Space-borne Methane IPDA Lidar Mission MERLIN

The CNES/DLR project MERLIN is a future IPDA lidar satellite mission that aims at measuring methane dry-air mixing ratio columns (XCH4) in order to improve surface flux estimates of this key greenhouse gas. To reach a 1 % relative random error on XCH4 measurements, MERLIN signal processing performs an averaging of data over 50 km along the satellite trajectory. This article discusses how to process this horizontal averaging in order to avoid the bias caused by the non-linearity of the measurement equation with measurements affected by random noise and horizontal geophysical variability. Three averaging schemes are presented: averaging of columns of XCH4, averaging of columns of Differential Absorption Optical Depth (DAOD) and averaging of signals. The three schemes are affected both by statistical and geophysical biases that are discussed and compared and correction algorithms are developed for the three schemes. These algorithms are tested and their biases are compared on modeled scenes from real satellite data. To achieve the accuracy requirements that are limited to 0.2 % relative systematic error (for a reference value of 1780 ppb), we recommend performing the averaging of signals corrected from the statistical bias due to the measurement noise and from the geophysical bias mainly due to variations of methane optical depth and surface reflectivity along the averaging track. The proposed method is compliant with the mission relative systematic error requirements dedicated to averaging algorithms of 0.07 % (± 1 ppb for XCH4 = 1780 ppb) for all tested scenes and all tested ground reflectivity values.